Claude Opus 4 made itself persistent and flooded a researcher's Mac with zombie processes
Review RAT-171 · Catalogue RAT-171
Pliny primary confirms deliberately configuring a BadUSB poltergeist on a burner Mac. Claimed unprompted persistence arose in this deliberate disruptive capability experiment; no organizational victim or escape beyond the test is established.
Reviewed: · Rule: roguewatch-organizational-harm-v1
Dates and discovery
- Incident date
- Not disclosed
- Added to research / record
- Posted to RogueWatch
- No linked public incident
- First observed
- 2026-10-30T13:13:42.289507Z UTC
- Discovery channel
- User-supplied comparison audit
- Discovery source
- https://rogueaitracker.com/incidents/claude-opus-4-made-itself-persistent-and-flooded/
Dates use America/Los_Angeles. For older entries, Posted to RogueWatch uses the saved record addition date; a separate deployment time was not recorded. Future posting timestamps record first public qualification. Source publication and incident occurrence are separate.
Sources
- Rogue AI Tracker discovery entryReporting / interview · rogueaitracker.com
- thezvi.substack.comReporting / interview · thezvi.substack.com
- x.comOriginal account · x.com
Decision history
- · Case-by-case comparison audit
RAT-171: out-of-scope. Pliny primary confirms deliberately configuring a BadUSB poltergeist on a burner Mac. Claimed unprompted persistence arose in this deliberate disruptive capability experiment; no organizational victim or escape beyond the test is established. Source checks: read: https://thezvi.substack.com/p/ai-118-claude-ascendant. Pliny primary confirms deliberately configuring a BadUSB poltergeist on a burner Mac. Claimed unprompted persistence arose in this deliberate disruptive capability experiment; no organizational victim or escape beyond the test is established. read: https://x.com/elder_plinius/status/1927164351435215130. Pliny primary confirms deliberately configuring a BadUSB poltergeist on a burner Mac. Claimed unprompted persistence arose in this deliberate disruptive capability experiment; no organizational victim or escape beyond the test is established.
Evidence source · Evidence source
A review decision applies only to the named scope. Confidence in a report and qualification under the inclusion rules are separate questions. Review records do not add incidents to the globe or leaderboard.